passing on poverty

Working Paper May 2011 No. 192

Chronic Poverty Research Centre

ISBN: 978-1-906433-64-2 www.chronicpoverty.org

What is Chronic Poverty?

The distinguishing feature of chronic poverty is extended duration in absolute poverty.

Therefore, chronically poor people always, or usually, live below a poverty line, which is normally defined in terms of a money indicator (e.g. consumption, income, etc.), but could also be defined in terms of wider or subjective aspects of deprivation.

This is different from the transitorily poor, who move in and out of poverty, or only occasionally fall below the poverty line.

Passing on poverty: the

intergenerational transmission

of well-being and ill-being in

rural Bangladesh

Peter Davis

Social Development Research Initiative 4 Victoria Place Bath BA2 5EY UK

Passing on poverty: the intergenerational transmission of well-being and ill-being in rural Bangladesh

2

Abstract

The paper presents findings on the intergenerational transmission of poverty from life

histories conducted in rural Bangladesh as part of a study of poverty dynamics and the

impact of development interventions. The life histories provide evidence for a relatively high

prevalence of intergenerational transmission of poverty (and of wellbeing). The means by

which poverty is passed on from parents to children is explored, painting a picture of poor

families being more exposed to, and more damaged by, a relatively small number of

categories of shocks and downward pressures, such as those caused by illness, crop

damage and floods, and with lasting effects often carried to the next generation. In addition,

the reduced ability of many poor people to benefit from a similarly small set of opportunity

types is also frequently passed on to the next generation. This is because the capacity to

seize opportunities is linked to tangible assets such as land and livestock, and intangible

assets, such as those that facilitate entry to salaried work. Poor endowments of these assets

tend to continue to the next generation. Some who benefit from success in business or

salaried work do manage to escape the poverty of their parents, but these people are the

fortunate few. They are also more likely to live in areas closer to the metropolitan centres.

Even while significant progress in reducing poverty has been made in rural Bangladesh in

recent years, it seems likely that tomorrow‟s poor will continue to be the children of today‟s

poor, unless more is done to ameliorate these socio-economic processes that allow poverty

to be passed from one generation to the next.

Keywords: intergenerational transmission of poverty, rural Bangladesh, poverty dynamics,

assets

Acknowledgements

This research was funded by the Chronic Poverty Research Centre. It builds on collaborative

work funded by the Department for International Development through the Economic and

Social Research Council of the United Kingdom, HarvestPlus, and the World Bank. I

acknowledge helpful discussions with, and comments from, Bob Baulch, Graham Brown,

Agnes Quisumbing and Sam Hickey. I also thank Zahidul Hassan and Md. Zobair of Data

Analysis and Technical Assistance Ltd (DATA) for managing the fieldwork team so efficiently,

and Biswas Akhter, Rafiqul Haque (Shawpon), Dilara Hasin, Safia Satter (Sonia), Md. Abdul

Aziz and Anowara Begum (Nupur) for excellent research assistance.

Peter Davis is a development sociologist based in Bath, UK. Trained as an agricultural

scientist, Peter also has a PhD in Development Sociology from the University of Bath where

he was a lecturer for ten years. Specialising in mixed methods approaches to poverty

research, Peter now dedicates himself to social development research and training with

SDRI and other organisations.


3

Email: [email protected]

This document is an output from the Chronic Poverty Research Centre (CPRC) which is

funded by UKaid from the UK Department for International Development (DFID) for the

benefit of developing countries. The views expressed are not necessarily those of DFID. The

CPRC gratefully acknowledges DFID’s support.

mailto:[email protected]


4

Contents

1 Introduction ..................................................................................................................... 5

2 Methods ........................................................................................................................... 6

3 Exploring the transmission of poverty and wellbeing .................................................11

3.1 Exploring the extent of transmission of poverty or wellbeing across generations ...................... 11

3.2 Exploring means of transmission of poverty: how disadvantage is passed on .......................... 13

3.3 Exploring means of transmission of wellbeing: how opportunity is passed on........................... 21

4 Discussion ......................................................................................................................27


5

1 Introduction

In this paper I draw from 293 life-history interviews conducted in rural Bangladesh in 2007 to

explore the extent and manner of the transmission of poverty or wellbeing from one

generation to the next. The life-history interviews were conducted as part of a unique mixed-

methods research project investigating poverty dynamics and the impact of development

interventions in rural Bangladesh (Davis and Baulch, 2010). This paper reports on the

intergenerational transmission of poverty, and also of wellbeing, by exploring and comparing

this life history information, in particular accounts of wellbeing in childhood compared with

assessments in adulthood.

Knowledge about the extent of, and the means by which, poverty or wellbeing is passed from

one generation to the next, is useful for shaping policies that can effectively reduce chronic

poverty and remove poverty traps. If, as is suggested by this paper, poverty in childhood

(which usually reflects parents‟ life conditions) frequently persists into adulthood in rural

Bangladesh, then policies aimed at reducing long-term poverty and removing poverty traps

should also attempt to interrupt this continuation of poverty from a parent to their children,

and therefore the intergenerational transmission of poverty. Since the long-term conditions

and processes that allow poverty to persist from childhood to adulthood tend also to be those

that allow poverty to transfer from one generation to another, learning about how childhood

poverty continues into adulthood also helps us understand how poverty is transferred from

one generation to the next. This paper attempts to present what we learned in this study

about the extent and means of the intergenerational transfer of poverty to help inform areas

of policy making which aim to reduce persistent chronic poverty.


6

2 Methods

For this paper, the life histories were examined to explore the long-term effects of poverty or

wellbeing in childhood and in some cases from reports of parents‟ life circumstances before

the respondent was born. In order to quantify some of the mobility patterns observed, we

were able to distinguish four groups of research participants: those poor in childhood but not-

poor in 2007 (move up); those not-poor in childhood but poor in 2007 (move down); those

poor both in childhood and in 2007 (chronically poor); and those not-poor both in childhood

and in 2007 (chronically not-poor).

Childhood conditions were based on qualitative retrospective descriptions of life

circumstances from the life history interviews; and conditions in 2006-2007 were based on

two independent assessments for each individual: one derived from per capita monthly

household expenditure from household survey data, compared with the national poverty line,

and the other based on wellbeing levels assigned by researchers after the separate life

history interview. This latter assessment was also crosschecked in focus group discussions

among local people. Where these expenditure-based and wellbeing-level-based

assessments differed, the cases were excluded from the analysis for this paper.1 This left a

sample of 160 life histories for this four-way classification, from a total of 293.

In the life histories, reports of childhood poverty nearly always reflected parents‟ poor life

circumstances and in only two out of 58 cases did these intergenerationally poor people

report life circumstances that reflected a temporary escape from poverty at some time

between a poor childhood and below poverty-line circumstances in 2007.

This paper is one of many reporting from this overall study.2 For the life-history component of

the study the author led a team of researchers from DATA Bangladesh3 in conducting the

293 life-history interviews in the eight districts in Table 1 (below) between April and October

2007. The households were selected as a subsample of a CPRC-IFPRI-DATA longitudinal

1 For an analysis of the reasons for differences between the expenditure-based and wellbeing-level-based

assessments see Davis and Baulch (2009). When all 293 life histories are analysed the findings presented in this paper are not substantially different. See page 6 for more discussion of the effects of excluding 133 life histories from this analysis.

2 See also Davis and Baulch (2009, 2010); Baulch (2010); Kumar and Quisumbing (2009); Ahmed (2005);

Hallman et al. (2007); Zeller et al. (2001).

3 Data Analysis and Technical Assistance Ltd. (DATA) is a research consultancy firm based in Dhaka with well-

2 See also Davis and Baulch (2009, 2010); Baulch (2010); Kumar and Quisumbing (2009); Ahmed (2005);

Hallman et al. (2007); Zeller et al. (2001).

3 Data Analysis and Technical Assistance Ltd. (DATA) is a research consultancy firm based in Dhaka with well-

established expertise in conducting large-scale social surveys and other research activities.


7

study of poverty dynamics in rural Bangladesh, which covered a total of 2,152 households

(1,907 original households) from 14 districts.4

Also as part of the overall study, three intervention types – microfinance, educational

transfers and agricultural technologies – corresponded with initial evaluation studies in the

quantitative survey, which were initially conducted separately, with baselines in 1994 for a

microfinance study, 1996 for an agricultural technology study, and 2000 for an educational

transfers study. These interventions are described in Zeller et al. (2001), Hallman et al.

(2007), Ahmed (2005), Kumar and Quisumbing (2009), Baulch (2010) and Davis and Baulch

(2009, 2010).

Households from the three initial evaluation studies were combined in 2006-2007 for three-

phases of sequenced qualitative-quantitative-qualitative research, to allow analysis of an

entire panel of 2,152 households, as if they had been part of the same longitudinal study.

This was possible because the initial studies had used similar survey methods, all

administered by DATA Bangladesh, who also conducted the 2006-2007 fieldwork. 5

For the subsample of life-history interviews (phase 3 of the 2006-2007 research), eight of the

original 14 districts were selected in such a way that a range of geographic and agricultural

conditions typical of rural Bangladesh were represented. Sites were selected across the

initial evaluation studies, and two villages per site6 were selected in different unions.7 In each

site, 20 households were selected from the original panel, across these two villages. Five

4 The dataset from this study is available publicly at: http://www.ifpri.org/dataset/chronic-poverty-and-long-term-

impact-study-bangladesh

5 The 2006-2007 study aimed to integrate and sequence quantitative and qualitative methods, in three

phases: Phase I involved focus-group discussions with four groups (of poor and better-off women, plus poor and better-off men) in each village. The focus groups aimed to elicit perceptions of changes, group members‟ perceptions of the interventions under study, and the degree to which these interventions affected people‟s lives (compared to other events in the community). Phase II was a quantitative survey of the original households and new households that had split off from the original households but remained in the same district. The household survey took place from November 2006 to February 2007, the same agricultural season as the original surveys, with multi-topic questionnaires designed to be comparable across sites and with the original questionnaires from the evaluation studies. Phase III consisted of a qualitative study based on life histories of 293 men and women in 161 selected households in 8 of the districts in the original quantitative study. The aim of this phase was to understand the processes and contexts which influence individual and household livelihood trajectories. Fieldwork for this final phase of the study was undertaken between March and October 2007.

6 „Sites‟ refer to districts in all cases except in Mymensingh and Kishoreganj districts where the „site‟ and the two

selected villages spanned the district boundary.

7 In rural Bangladesh a union is the administrative area overseen by the lowest level of local government, the

Union Parishad (Council), and usually includes several villages.

http://www.ifpri.org/dataset/chronic-poverty-and-long-term-impact-study-bangladesh

http://www.ifpri.org/dataset/chronic-poverty-and-long-term-impact-study-bangladesh


8

households were randomly selected from each of the four poverty-transition categories8

based on per capita household expenditure calculated from the quantitative survey

comparing data from the baselines (1994,1996, 2000) and the 2007 data (see Davis and

Baulch (2010) for a more detailed explanation of these selection methods).

Table 1. Locations of the life history research villages

Original Intervention Type District Number of Villages

Number of Life Histories Conducted

Microfinance (MF) Manikganj 2 36

Kurigram 2 39

Educational transfers (ET)

Nilphamari 2 38

Tangail 2 39

Cox‟s Bazar 2 32

Agricultural technology (AT): household-based fish farming

Mymensingh 1

18

Kishoreganj 1 19

Agricultural technology (AT): group-based fish farming

Jessore 2 36

Agricultural technology (AT): improved vegetables Manikganj 2 36

Total 16 293

In the life history interviews, one man and one woman were interviewed separately in each

household. Research participants were often husband and wife, but in some cases, such as

when a partner had died, we interviewed one parent and their son or daughter. When a

household member wasn‟t available – which was more often a man than a woman,

especially during the main April-May rice harvest – or when there was only one adult

household member, we conducted one life history interview in the household. In the end we

conducted 293 life history interviews in 161 households (133 men and 160 women) in 16

villages and eight districts. Interviewing two adult household members allowed immediate

cross-checking, gave a gendered perspective, an alternative view of household dynamics,

and allowed a mixed-sex team to work effectively, with men usually interviewing men and

women interviewing women.

On the same day of each life-history interview, interviewers wrote up the interview in Bengali

in a format which we had prepared in an initial training workshop and refined in field

discussions. Interviewers also wrote fieldwork diaries containing reflective impressions and

lessons-learned about methods, in addition to the more formally-agreed write-up of the life

8 These categories are chronically not-poor, chronically poor, move up and move down with respect to the per

capita expenditure national poverty line at the time of the baseline and 2006-2007 surveys.


9

history. These diaries were translated and became a part of the qualitative data set. When

with the team, the author participated in interviews with men or women.

Interviews and focus-group discussions were recorded, with permission of the research

participants, with small unobtrusive digital voice recorders. The research team did not

attempt to write full transcripts, but digital recordings were used for checking back on

interviews for the initial same-day write-up of the life-history narrative in Bengali, for later

analysis, and for the final anonymised write-up in English.9

Each life history was written as a chronological account of life events, identifying causal

mechanisms and drawing from discussions which had encouraged counterfactual thinking.

The aim was to produce, as accurately as possible, the participant‟s perspective on his or her

life trajectory, the causes behind improvement or decline in wellbeing, and how life could

have been if the events that emerged – both positive and negative – had not occurred.

The research team spent about two weeks in each of the eight sites during the life-history

phase of the research and later revisited most of the households to check and discuss texts

and diagrams with participants. As the principal analyst of the findings, the author visited and

discussed the research with participants in every household in the life history study. We also

arranged a time for „knowledgeable people‟ to attend a focus-group discussion on the last

day of our research in the village. This meeting was usually held in a school building or in a

near village leader‟s house. We tried to have a Union Parishad (Union Council) member and

a number of elderly people in attendance.10

During the life-history interviews we used national and local historical markers, such as the

1971 war of independence or the 1988 floods, to pinpoint years of events described by the

research participants. As an interview progressed, a chronological timeline of life events was

built up. At the end of a life history interview, the researcher who had facilitated the interview,

in discussion with the interviewee, drew a line depicting ups and downs in wellbeing in a

trajectory using this timeline of events. After this the interviewer reviewed the main points of

the life trajectory and asked the interviewee to identify and rank the three or four most

important sources of opportunity that had made the most difference for his or her long-term

wellbeing. Similarly the three or four most important sources of downward pressure were

also identified and ranked. These choices were recorded and used to generate frequencies

of upward opportunities and downward pressures, as shown in Tables 5 and 6. The other

researcher present at the interview wrote the narrative-based life history from notes taken

9 Examples of these life histories are provided at http://www.sdri.org.uk/bangladesh.asp.

10 In this paper focus-group discussions refer to these exercises rather than a separate set of 116 focus-group

discussions conducted in 2006 with findings reported in Davis (2007).


10

during the interview and the recording of the interview. Only two interviews were conducted

per day by each pair of researchers, to allow time for the diagram to be finalised and the

interview to be written up in Bengali on the same day.

The author then traced, translated and anonymised the life history diagrams and made these

available with the final anonymised, translated and edited narrative life histories. The level of

wellbeing (or „life condition‟ – obosta in Bengali) at different points in the life trajectory was

indicated on the diagrams using a scale of one to five, using the categories described in

Table 2 below, based on life-conditions described by the research participant in the interview,

and cross-checked in the focus groups.

Table 2. Qualitative wellbeing levels for individuals

Level English Bengali Guideline

1

Very poor or destitute

khub gorib, na keye chole

Suffering tangible harm to health because of poverty, generally due to insufficient food. Tend to be landless or near landless.

2 Poor gorib Very vulnerable but eating reasonably well. Could easily move into level 1 due to an unexceptional shock. If land is owned, it is equivalent to less than an acre for a medium-sized household.

3 Medium madhom A common shock would not result in tangible harm or going without food. Have household assets, or generate household income, equivalent to between one and two acres of land for a medium sized household.

4 Rich dhoni Hold household assets or generate household income equivalent to that generated by two to ten acres for a medium-sized household.

5 Very rich khub dhoni Hold household assets or generate household income equivalent to that generated by ten acres or more for a medium-sized household.


11

3 Exploring the transmission of poverty and wellbeing

3.1 Exploring the extent of transmission of poverty or wellbeing across generations

A relatively high level of transmission of poverty and wellbeing across generations appears

from the analysis of life histories when compared to wellbeing status in childhood and

adulthood. 63 percent of those poor in childhood remained poor in adulthood, and 74 percent

of those not-poor in childhood were also not-poor in adulthood. This suggests that both

advantage and disadvantage are passed on to a relatively high extent in rural Bangladesh,

but also reflects the overall trend of poverty reduction observed by others (Sen and Hulme,

2006).

These findings do need some qualification. First, 133 cases were excluded where the

qualitative and quantitative assessment for 2006-2007 differed. One of the main reasons for

this difference in assessment is the bluntness of both techniques in assigning wellbeing

status when people are close to the poverty line. But this may lead to an underestimation of

upward mobility overall, as a number of respondents who may have just crossed the poverty

line during their lives may have been excluded. In addition we are aware that our qualitative

assessments of wellbeing may underestimate upward mobility while the quantitative

assessments probably overestimate this (see Davis and Baulch (2009) for a discussion of

these differences).

Table 3. Transition matrix for childhood to adult socioeconomic mobility

poor in 2007 (percent) not poor in 2007 (percent)

poor in childhood (percent) 37 22

not poor in childhood (percent) 11 30

The life history information also suggests that the intergenerational transfer of poverty

increases with depth of poverty in childhood. Children who were so poor in childhood that

they regularly went without food (level one in our qualitative scale) were much more likely to

be in poverty in adulthood, with about 87 percent of this group of very poor children

remaining poor in adulthood.

There also seems to be some evidence from our study for persistence in intergenerational

transmission of poverty for the poorest in the post-1971 years, even though overall poverty

rates have clearly declined. For people born before 1971, 57 percent of those poor in

childhood remained poor in 2007. For those born after 1971, 79 percent of those poor in

childhood remained poor in 2007. We need to keep in mind that at the time of independence


12

in Bangladesh, headcount poverty rate was much higher. Also because the final year of

assessment was the same for both cohorts, older respondents had much more time to move

out of poverty in a context of long-term reduction in headcount poverty rates. Nevertheless, it

is worrying that for the smaller proportion of the overall population that were poor in

childhood in more recent times, prospects for socioeconomic mobility still seem to remain

restricted in rural areas. This could lead to an excluded group of poor people at the bottom of

the socio-economic distribution who may have declining prospects of escaping poverty in the

future. Again this observation will need confirmation with research using larger, more

representative samples. Nevertheless, this study raises concerns for the long term prospects

of the poorest – and their children – in rural Bangladesh.

When we compare respondents living close to Dhaka (in the four Manikganj villages) with

those remote from a metropolitan centre, we find that a larger proportion of respondents (59

percent) have moved out of their childhood poverty in the sites close to Dhaka, even though

on average the proportion of chronically not-poor respondents was lower in these villages

(8.9 percent). This indicates a much more socio-economically mobile population in these

areas. In contrast, more remote areas had a higher proportion of the chronically non-poor but

were also home to 69 percent of the move-down category in our sample. The increased

upward mobility in the villages in Manikganj District, close to Dhaka, is probably associated

with greater opportunities for business and salaried work. It may also be linked to more

rapidly increasing land values, and better access to health services, NGO programmes,

schools and markets.

Table 4. Differences between remote and non remote sites

not remote percent remote percent

intergenerational move down 31.3 68.8

intergenerational move up 59.3 40.7

intergenerational not poor 8.9 91.1

intergenerational poor 25.5 74.5

Our findings also suggest that the intergenerational transmission of poverty may be more

severe for women than for men. For men in our study, 58 percent of those poor in childhood

remained poor in 2007, whereas for women, 66 percent of those poor in childhood were poor

in 2007. However this finding also needs to be qualified by the fact that female respondents

were on average eight years younger than male respondents, so a cohort effect may be

complicating this observation.


13

3.2 Exploring means of transmission of poverty: how disadvantage is passed on

While case-based qualitative research suffers from the disadvantage of having small

numbers of cases, it benefits from the thick descriptions of each case that are possible,

allowing a unique perspective on the processes that lie behind the frequencies presented

above.

In order to understand the processes which allow poverty to be passed on from one

generation to the next, we need to examine both the transmission of disadvantage as well as

that of advantage. Poor children fail to escape poverty in adulthood due to a surplus of

constraints and downward pressures, and also due to a dearth of upward opportunities or

lack of capacity to gain from opportunities. In this section I first explore patterns of downward

pressures that affected those people who remained poor from childhood into adulthood and

compare these with those who, in adulthood, had managed to escape their childhood

poverty. In the following section I then explore what we learned about the distribution of

various forms of upward opportunity. The combination of both helps to explain the

persistence of poverty across generations, as is illustrated by the selected cases presented

below.

Table 5 provides a comparison between the rankings of the most frequently reported causes

of downward pressure from the life history interviews as cited in the entire set of 293 life

histories, and then across the intergenerational move down, move up, chronically poor and

chronic not-poor groups (which total 160 respondents due to the exclusion, in this analysis,

of cases where the qualitative and quantitative assessments of wellbeing in 2007 disagreed).

People in the move-down and move-up categories are those that declined into, or escaped,

poverty between childhood and 2007.


14

Table 5. Types of downward pressure

percent of all life histories

move down percent (of 17)

move up percent (of 35)

chronically poor percent (of 59)

not poor percent (of 49)

illness or injury 75.1 76.5 68.6 76.3 75.5

dowry or marriage 38.9 17.6 45.7 32.2 38.8

death of family member 33.1 35.3 17.1 28.8 36.7

household or property division 21.5 17.6 22.9 27.1 24.5

theft or cheating 20.1 11.8 17.1 18.6 38.8

litigation 19.1 47.1 34.3 5.1 34.7

floods, cyclones or storms 17.1 29.4 11.4 20.3 16.3

crop damage 14.3 23.5 14.3 5.1 24.5

violence, conflict, or physical insecurity 14.3 11.8 8.6 16.9 18.4

family or village disputes 9.9 17.6 20.0 6.8 10.2

illness or death of livestock 9.6 0.0 11.4 10.2 4.1

disability 9.2 5.9 2.9 11.9 10.2

unemployment or low income 8.9 11.8 8.6 11.9 6.1

business loss 8.5 11.8 14.3 5.1 14.3

sale or mortgage of land or house 8.5 17.6 5.7 8.5 6.1

debt 7.2 0.0 11.4 10.2 4.1

supernatural causes or superstition 6.8 5.9 0.0 10.2 4.1

divorce or abandonment 6.5 0.0 2.9 5.1 6.1

migration 6.5 5.9 2.9 5.1 16.3

extortion, corruption, or harassment 5.8 11.8 14.3 0.0 10.2

lack of food 4.1 0.0 2.9 10.2 0.0

education or other children's expenses 3.8 5.9 8.6 1.7 6.1

fire 3.4 0.0 0.0 5.1 8.2

3.2.1 Illness

Table 5 shows that illness or injury was the most frequently cited downward pressure across

all these groups, but was least frequent in the intergenerational „move up‟ and „chronic not-

poor‟ groups. This suggests that illness and injury is both a major cause of impoverishment

over the long term, and that this extends into the intergenerational transmission of poverty.11

I use details from a particular life history below to illustrate mechanisms that were commonly

seen to allow illness to be a cause of intergenerational poverty. Ali‟s life trajectory (See

11 The poorest also tend to be more exposed to and damaged by illnesses and injuries. See Davis (2011) for

more analysis from this study of the skewed exposure of poorer people to illness.


15

Figure 1) shows how poverty can be exacerbated by illness, which in turn makes people

more susceptible to other crises later in their lives, and ultimately causes the

intergenerational transmission of poverty. For Ali, it is possible to trace the impact of a series

of health related misfortunes across three generations from his parents, through his, and on

to his children‟s current low level of wellbeing.

Figure 1: Ali: 45 year-old man, Tangail District

Ali is a 45 year-old man living in Tangail District, about three hours drive north of Dhaka. His

life-trajectory improved slightly over the long-term, but not enough to rise above the „poor‟

level 2, which is about the level of the poverty line.12 The death of Ali‟s father when Ali was an

infant can be seen as a large shock which has undermined Ali‟s ability to improve his life,

and in turn has led to disadvantage for his children. Opportunities to improve have not been

absent; and ongoing constraints present throughout his family history have restricted his

ability to convert these opportunities into life improvement.

12 Ali‟s household per capita expenditure rose from about half the poverty line in 2000, to just over it in 2007, in

our quantitative assessment. However with household instability and health-related expenses, including Ali‟s ongoing mental illness, it is unlikely that the observed rise in household expenditure accurately reflected a corresponding rise in wellbeing for the same period. See Davis and Baulch (2009) for a discussion of disagreements of quantitative- and qualitative-based poverty assessment.


16

Box 1: Ali, 45

Ali‟s father had been an agricultural worker and earlier in his life had owned about 5 bighas of land.13 He died from TB when Ali was a year old in 1963. Before he died the family sold land to pay for his treatment leaving 2 – 2.5 bighas for the family to live on. Ali and his brothers all missed out on learning to read or write and other forms of elementary education as they were forced from a young age to work as day labourers to keep the family going.

Ali started work in 1973, when he was about 11. He tended other people‟s cows and earned about Tk 700 – 800 at the beginning, and was fed, housed and clothed. He did this work for seven years and by the end of this time he was earning about Tk 4500 annually. He worked for four different neighbours. In 1980, at the age of 18, Ali returned home and started work as a regular agricultural day labourer.

Ali‟s brothers formed separate households in the same neighbourhood by 1985 whilst Ali stayed with his mother. His brothers believed their situation would improve if they were by themselves and so they chose to be separated. Ali said to us that his wellbeing was not affected much when his brothers were separated, as at the time he was working by himself. When his brothers separated, their 120 decimals of paternal land was split three ways, and Ali received 40 decimals - out of this, 16 decimals was public land (khas jomi).

Ali was married in 1986, a daughter was born in 1988, but his first son died as an infant from typhoid in 1990. The baby was treated by the traditional healer (kobiraj), which cost TK 400-500, then he was taken to the local government hospital where he died the same day. The death of their son affected Ali and Hasina profoundly, Ali became mentally ill and Hasina suffered from depression.

Due to Ali‟s mental illness he spent several years wandering and sleeping rough. Ali said that he was possessed by a supernatural being (jin). He wandered around the district being fed by others, sleeping rough and being treated as a madman (pagol). He only came home from time to time. After seven years Ali began to suffer from abdominal pain for which he received extensive hospital treatment which cost a total of Tk 20,000. This money was raised through the sale of pineapples and with help from his father-in-law.

During this time his wife, Hasina, had to try to survive without any income from Ali and as a result they were forced to mortgage land and take loans in order to live. In 1992, Hasina contracted a fever and a cough and was later diagnosed by a doctor in a nearby town as having tuberculosis. She was taken to the local government hospital for treatment and then was sent to a specialist TB clinic in a hospital near Tangail. The treatment cost Tk 1500 which they again raised with help from Ali's father-in-law. Hasina recovered after 6 months, but she had been pregnant with a daughter, born in 1993, who was born blind, partially deaf and with severe learning difficulties.

A second son was born in 1995, and another son in 2000. The older son now attends primary school and has received the Primary Education Stipend (cash-for-education) for the last three years. Their eldest daughter also receives a stipend for attending school. Most of this money is used to pay for the children‟s education and clothing expenses. Ali said that if he didn‟t receive this money they would have to cut back on food and eat only twice a day.

13 In this area there are 48 decimals per bigha or pakhi of land; there are 100 decimals in an acre of land. It is

difficult to determine whether Ali‟s father would have been classified as poor or not poor earlier in his life before his illness from the life-history interview. Five bighas of land was less productive in 1962 than in 2007, and an asset-based classification would need to account for this. Certainly once Ali‟s father became ill and land was sold to pay for treatment he would have been classified as poor using our qualitative criteria.


17

However, their disabled second daughter is unable to attend school due to her disabilities. Even though her condition is deteriorating, they have been unable to seek treatment for her because of the cost. They worry for her future as it is unlikely that they will be able to find a husband due to her disabilities.

In recent years Ali has had medication for his mental condition and has returned home and has been working as a day labourer, driving a van rickshaw and growing pineapples on their small plot of land. However without this string of health problems and other pressures, which can be traced back to vulnerability created by his father‟s illness and death, he and his family would be in stronger position today.

It is difficult to make linkages like this with absolute certainty; however the death of Ali‟s

father caused the sale of assets and loss of income which increased Ali‟s vulnerability for a

long time due to a resulting lack of coping resources. The „bad luck‟ experienced is an

outcome seen again and again in the life histories and this case is typical in as much as it

shows how idiosyncratic spates of bad luck commonly emerge due to the long-term

vulnerability caused by denuded coping resources. The observation that common, yet

idiosyncratic, „bad luck‟ is more frequently experienced by the chronically poor also

problematises the popular distinction between idiosyncratic and covariant shocks:

idiosyncratic shocks tend to be experienced disproportionately by the poorest and are

therefore patterned – as are covariant shocks.

3.2.2 Dowry

After illness and injury the next most frequently cited cause of downward pressure overall –

dowry and marriage expenses – seems to be a problem more frequently for those moving

out of intergenerational poverty than by those who are declining into it (and to a lesser extent

those remaining in it). This observation is difficult to interpret. An exploration of the smaller

number of „move down‟ respondents who cited dowry as a main downward pressure in their

lives shows that dowry and marriage did not feature as frequently because many of these

respondents were yet to face dowry for their daughters because their children were still

young. We also need to keep in mind that dowry has emerged as a cause of impoverishment

relatively recently in Bangladesh with reports of large dowries quite rare in the pre-1971

period. As a result the impoverishment caused by dowry is a more recent phenomenon.

However, for households who have girls of marriageable age, dowry is an important cause of

downward pressure for both rich and poor. Unfortunately marriage of daughters also often

corresponds at a time in life when a couple‟s parents are also elderly and demanding

financial resources for medical care. This can create a „double whammy‟ for some mid

generation people when they are forced to finance medical care for the older generation at

around the same time as financing marriage expenses for the younger generation.


18

3.2.3 Death of family members

Table 4 also suggests that circumstances around the death of family members are also a

means by which poverty is transferred from generation to generation. This is often passed on

due to assets being sold or loans taken to pay for medical care before death. It also leads to

withdrawal from school and depletion in „human capital‟, which could later undermine

prospects for improvement through business or salaried work.

Sonia‟s life trajectory (Figure 2) illustrates causal links that start with the death of a family

member – in this case her father – and leads to disadvantage that crosses generations.

Figure 2: Sonia: 35 year-old woman, Manikganj District


19

Box 2: Sonia, 35

Sonia (35) lives with her husband, Nurul (65), and her one son (11) and one daughter (6). Her son is in class 5 at a government primary school and her daughter studies at a local mosque (gono shikkha karjokrom) (when they have completed this study, students are able to enter secondary schools in class 6). Sonia and Nurul own about 50 decimals of land (half of an acre) and Nurul works this land and has a seasonal crop business (he buys and sells molasses (kejuri gur) and onions and chillies when they are in season).

Sonia was born in 1972 in a village about 5 kilometres from where she now lives. Her father used to ply a boat and her mother was a housewife (grihini). Sonia was the youngest of four sisters and two brothers. Her father didn‟t have any land apart from where their house stood. The family survived on her father‟s income as a boatman, but at a low level (level 2) and her father died when she was 11 years old.

Before her father died, her three sisters were married and moved to their husband‟s households. After her father died, her family faced severe difficulties: her elder brother took a job in Dhaka in a small printing press business; her younger brother took up agricultural day labour (mathe din mojur); and her uncle (father‟s older brother or chacha) helped them with rice to eat. Also after her father died, her brother‟s earnings were not sufficient to pay for her continuing education and she had to leave school after class five and didn‟t do any more study. Sonia explained, “It was difficult enough provide food and clothes from my brothers‟ earnings, so how could I study? (bhaider takai poroner kapor ebong bhat jogano koshtokor chilo, ki bhabe porbo?)”.

In 1992, at twenty, Sonia was married for the first time; this was considered late for a girl not at school, but because she didn‟t have a father, they couldn‟t save for her dowry and marriage arrangements were delayed. Sonia‟s uncle and brothers arranged for her to be married with a boy who had worked as a caretaker and livestock tender (rakhal) in her uncle‟s house. However Sonia didn‟t want to accept (mene nite pare ni) this marriage for two reasons: first the boy had worked at her uncle‟s home and this meant that he was of lower status than she; second he had a disability – he couldn‟t do anything with his left arm. Her family paid a small, Tk. 3,000, dowry for this wedding. But the day after the wedding, Sonia left her husband‟s house and returned to her mother‟s house and didn‟t go back to her husband. A month later she divorced (talak) him and the dowry was not returned to her family.

In 1992 her mother died. Following this, Sonia stayed in the same household as her brother and his wife and children. However, in 1993 Sonia‟s elder sister, Nurjahan, also died, suddenly from diarrhoea – it was probably cholera but she died on the way to hospital on the van gari (flat-decked cycle rickshaw) and was never diagnosed. Nurjahan‟s mother had given Nurjahan a cow as a gift and when she became ill, her husband sold the cow to pay for her medical treatment. In the end they were unable to save her. One month after this Sonia was married to Nurul, her sister‟s widower. Sonia was not happy with this marriage but she had to accept the decision of her two brothers and uncle. All along she has felt that there was too great an age difference as Nurul is 30 years her senior.

When Sonia married Nurul, he owned 50 decimals of cultivatable land, two goats and two sheep. Nurul worked some of the time on other people‟s land as a day labourer. He also traded agricultural produce on a small scale when they were in season. He bought and sold onions and chillies in the summer time, and molasses (kejur ghur) during the winter.

In 1996 her first son was born, and in 2001 a daughter followed. However, Nurul always struggled with not having enough capital to make his business successful. Usually he borrowed money from his relatives, often from his maternal cousin‟s husband (mamato boner shami). When he had finished his trading, he returned the capital to his cousin‟s husband without paying any interest and kept the profit himself.


20

Currently her eldest son receives Tk. 100 per month from the cash-for-education programme but her daughter doesn‟t receive any aid from her school. Sonia also sells chicken and duck eggs to pay for her children‟s education costs: pens, pencils, exercise books and school uniforms.

Sonia has also been a member of an NGO cooperative group for the last year. Every week she saves 5 to 10 taka. Last November she took out a Tk. 1000 loan in her name to contribute money for her husband‟s business. Every week she pays back 25 taka to the NGO co-op and will pay 44 instalments along with Tk 5-10 per week in savings. She gets the instalments from her husband‟s business although sometimes she has to borrow money from her relatives to make the payments, which are then paid back when the business yields some profit. At the time of the interview (April), it was a bit difficult to make the instalments. In this area the bangla month of boishak (April-May) is one of the two periods of want in the village. At this time the boro (dry season) harvest is not yet in and there is low demand for day labour.

In 2006 Sonia had a stomach ulcer. To treat this they sold a goat for Tk. 2000 at the time of Kurbani Eid. Even though she has few children, it seems that the conditions in her household are slowly declining. There are a number of indicators that suggest this decline:

- her husband is getting old and cannot work as hard as he used to,

- her children have not started working yet and have a number of costs

associated with their education,

- Sonia does not earn much for herself and has outstanding loans –

only half paid back so far.

As Sonia discussed her future prospects, she said she would like her children to stay in school and at least get more education than she had. She would also like her daughter not to face the same problems when she is older. However, Sonia and Nurul have limited resources and will probably also face difficulties when their daughter is to be married.

Sonia‟s life trajectory illustrates the way that both downward pressures and restricted

opportunities combine to perpetuate poverty over the long term. The negative effects of her

father‟s death when she was young created a situation where her marriage prospects were

limited, and this resulted in one failed marriage and another unhappy one. She and her

husband have few assets and limited education, and with Nurul‟s advancing years, there is

little scope for improvement in their lives, and a high likelihood of a number of impending

hardships for them and their children.

3.2.4 Other forms of downward pressure

While illness and injury, dowry and marriage, and deaths of family members were the most

frequently cited downward pressures over the entire set of life histories, poorer households

tended to cite illness and injury and deaths of family members more frequently, and dowry

and marriage slightly less frequently than respondents above the poverty line in 2007. Of the

range of downward pressures experienced, frequencies of citation varied across the four

intergenerational move up, move down, chronically poor, and chronically not poor groups,

but the patterns observed are quite difficult to interpret (Table 5). There is an overall pattern

which shows that the move up category tended to cite pressures such as the death of family

members; floods, cyclones and storms; violence, conflict and physical insecurity; disability;

and sale or mortgage of land or house, less frequently than other groups. They also cited


21

dowry and marriage slightly more frequently, indicating that while dowry is a downward

pressure for all, those moving out of poverty may be investing more in dowry than other

groups. However, it may also be that those who are moving up out of poverty may be at a

stage in their life-cycle where dowries are being paid for daughters, so lessons from these

patterns are not clear. In contrast individuals moving down across generations (from

childhood into adulthood) tended to cite illness and injury; litigation; floods, cyclones and

storms; unemployment and low income; and sale or mortgage of land or house, more

frequently than other groups.

The relatively small number of cases analysed and the qualitative nature of these

assessments make generalisations from these observations difficult. Certainly the frequent

citation of litigation across all categories - except the chronically poor who presumably are

less likely to have the resources to pursue litigation - is a cause for concern, and the higher

frequency of citation of sale or mortgage of land or houses among the move-down category

suggests that holding on to assets is key in preventing decline.

3.3 Exploring means of transmission of wellbeing: how opportunity is passed on

In this section I draw from our life histories to explore the processes which allow

opportunities to be passed on from one generation to another.

Table 6 provides frequencies of cited sources of life-improving opportunity across the entire

set of life histories and the intergenerational mobility groups. This shows that business, land-,

and livestock-related opportunities were the most frequently cited overall. This again

highlights the important role of tangible assets in providing life-improving opportunities. When

we compare these across respondents from our four mobility categories (poor in childhood

and adulthood; not poor in childhood and adulthood; not poor in childhood but poor in

adulthood; and poor in childhood but not in adulthood) we observe different frequency

patterns. It seems that business income and land accumulation were frequently important for

those escaping from their parent‟s poverty. It seems also that having sons or daughters

working was a frequent source of opportunity for all categories, except those moving into

poverty from a non-poor childhood.


22

Table 6: Types of upward opportunity

percent of all life histories (of 293)

move down percent (of 17)

move up percent (of 35)

chronically poor percent (of 59)

chronic not poor percent (of 49)

business 46.1 47.1 68.6 37.3 38.8

land 38.6 29.4 48.6 28.8 32.7

livestock 27.3 11.8 22.9 27.1 14.3

sons and daughters working 24.2 0.0 22.9 35.6 26.5

agriculture or pisciculture 23.5 11.8 17.1 16.9 32.7

help family and inheritance 22.2 23.5 14.3 18.6 28.6

day labour 18.1 11.8 5.7 25.4 10.2

benefits from official programmes 16.7 23.5 5.7 25.4 6.1

loans 16.0 23.5 25.7 18.6 8.2

salaried work 14.3 11.8 22.9 11.9 12.2

dowry receipt or favourable marriage 11.6 11.8 22.9 6.8 4.1

household or property division 8.5 0.0 5.7 8.5 14.3

domestic labour migration 7.2 5.9 2.9 10.2 2.0

house 4.1 5.9 2.9 5.1 0.0

irrigation 3.8 5.9 5.7 0.0 10.2

united family 3.4 11.8 0.0 3.4 2.0

Day labour also seems to be important for survival. This was suggested by 25 percent of the

chronically poor citing it as a source of opportunity, but not as an opportunity for upward

mobility. Only 6 percent of those moving up cited day labour as a significant source of

opportunity.

Official programmes (such as the Vulnerable Groups Development Programme) seem to be

fairly well targeted with the highest percentages of citations of these programmes coming

from the chronically poor or those moving down. However, these frequencies also suggest

that official programmes have a more protective role, rather than as a means of lifting people

out of chronic poverty.

These frequencies do suggest that salaried work is an effective poverty-exiting opportunity

with those moving up citing it nearly twice as often. Interestingly, the receipt of dowry also

seems to have been more frequently cited by this group.

Those who were not poor in childhood cited inheritance more frequently as a source of

opportunity; presumably their lack of poverty in childhood was reflected in assets owned and


23

passed on while those poor in childhood cited inheritance much less frequently, including

those who had managed to move up through other means.

Rubel‟s life trajectory (Figure 3) illustrates a number of these common causal mechanisms

showing how opportunities, such as salaried work and land accumulation fit into the wider

context of a struggle to break out of childhood poverty in a difficult environment. However,

Rubel achieved this with some education, and hard work in a job that takes him away from

home for most of the time. The income from a stable salaried job in Dhaka (his home is

about an hour and a half by road from Dhaka), coupled with his astute use of excess income

and loans to buy land and irrigation equipment near his house, has been important in this

story of upward mobility.

Figure 3: Rubel: A 47 year-old man, Manikganj District


24

Box 3: Rubel, 47

Rubel (47), his wife, Shireen (39), and their son (16), daughter (22) and granddaughter (1), live in the village of Narikelbari, in Manikganj District. Their daughter and granddaughter stay with them for eight months of each year, and stay the remaining four months in their daughter‟s husband‟s household. Their son is not married. Rubel works for a printing press in Dhaka and only returns home on Fridays, his one day off a week. Because of his work in Dhaka and slow accumulation of land, paid for from loans, salary and farming, Rubel has been able to enjoy an improving quality of life since the mid 1980s when he separated from his brothers after his father died. Even though Rubel‟s endowments of tangible assets were few, it seems that he benefitted from enough education, and coupled with willingness to work hard and astute investments he has been able to improve his life over the long term.

Rubel was born in 1960, the son of a small farmer and one of 7 children (3 boys, 4 girls, who are all still living). Rubel was the youngest son, with one daughter younger than him. He received a basic education, up to class 5; only one other son and daughter of his family went to school, and they only stayed till class 3. The rest of his family had no education at all. Rubel reported to us that at the time he was born, his family was very poor (khub obhabi chilo). Over the following years when he was a child, the costs of arranged marriages for his sisters and continued medical expenses for his father caused their family to further decline into poverty.

Rubel‟s father originally owned 22.5 bighas (7.5 acres) of land. When his eldest daughter was given in marriage in 1963 (aged 10) he sold three bighas to pay for the cost of the wedding; the only dowry given for this marriage was several items of jewellery. When Rubel‟s second sister was married in 1965, at 11 years old, items of jewellery were also given and two bighas of land were mortgaged to pay for these, raising the sum of TK 900. Additional money was also raised to pay for the wedding. A cash dowry was not provided for either marriage.

Over these years (between 1963-1969), mortgages were taken out over a further 15.5 bighas of land, to provide money for the ongoing living expenses, Rubel‟s sister‟s marriages and his father‟s medical expenses. It seems, therefore, that a period of decline was taking place early in Rubel‟s childhood due to illness and dowry and marriage costs.

In 1969 when Rubel was nine years old, his father was taken seriously ill and was unable to work or even leave their house. In the absence of anybody who was able to support them, Rubel pulled out of school after class five and assumed the responsibility of providing for the family by working their land, going to the bazaar, organising people to plant and harvest their rice and other work. Rubel‟s older brother, who had received no education, was also working, but in a jute mill away from home, and sent money home; his other brother could not help much.

Rubel‟s oldest brother was married in 1972, at 17 years old. The marriage was arranged by an uncle (chacha) because at the time his father was still ill. Before the marriage they had no money to pay for expenses so they mortgaged one bigha of land. The family received TK 1,700 by way of dowry. This was then all used to pay for wedding expenses. Items of jewellery valued at TK 600-700 were also given by the bride‟s family, but Rubel‟s father gave some jewellery (silver ear rings and necklace and a gold bracelet) in return. This meant that overall there was little financial gain to the family from the marriage. In 1976, the third daughter was married, at 19 years old, again without dowry, to a groom who also worked in the jute mill. Some jewellery was given by Rubel‟s brothers for this wedding. The family met the other costs of the wedding by selling crops.

After several years of working on their land and also working as a day labourer, in 1978 Rubel got a job working for a printing press in Dhaka. During the period from 1978 to 1980, he was regularly able to earn extra money by working overtime, and his financial situation improved steadily. Rubel‟s 2

nd brother was married in 1981. He had started working in 1976 with Rubel, on their land.


25

It was unclear from the interview whether the family received dowry from his bride‟s family. His youngest sister was married in 1982. About Tk 3500 of jewellery was agreed to be given and wedding expenses were Tk 3000. This money came from selling crops and from the wages of the three brothers. At this time the conditions of the family were not too bad (muta muti shotchol).

By 1980, Rubel and his brothers had amassed sufficient funds between them, from Rubel‟s job in the printing press and his eldest brother‟s job in a jute mill, to pay off all the mortgages previously taken out by their father. They also were able to buy a further 3 bighas of land from their uncle. However, the family land lay near a river, and over a few years leading to 1983, all the land and the houses on it had been eroded away and lost to the river. This event led to a second decline in their lives.

When Rubel was married in 1983, aged 23, his family arranged a dowry comprising a tape recorder valued at TK 1,700 from the bride‟s family. However, as had been the case in his brother‟s marriage, Rubel gave his wife a gift of jewellery valued at about Tk 2000 in return, meaning no real financial advantage accrued to the family from the marriage. Rubel and his wife, Shireen, continued living in the family home for some years after their marriage.

Just after he was married, Rubel spent TK 12,000, saved from his job, and purchased 12.5 decimals of land in Narikelbari, several miles away from his birthplace and original home, on which to build a new home. He did this because they had lost their land to the river and he had also fallen out with his oldest brother over sharing what remained of their family property. Later his other brother also split with his oldest brother and there was a quarrel over family property.

In 1986, Rubel‟s father fell ill again and died a short time later. Rubel remembers having paid medical costs amounting to TK 500-600 around this time but he was in Dhaka working when his father died. From this time on Rubel‟s mother stayed with her other sons, but from 1988, from the time of serious flooding, until her death in 2006, she stayed permanently with Rubel.

1986 also saw the birth of Rubel‟s first child, a daughter, but in the same year Rubel had TK 2,300 stolen by a pickpocket in Dhaka while shopping for provisions for the festival of Eid. He was greatly troubled by this loss, and being unable to provide properly for Eid. Even though he was eventually able to purchase the provisions he needed on account, he notes it as a low point. A similar loss of TK 3,500 also occurred in 2002 in Dhaka, this time during a mugging.

Rubel himself fell seriously ill in 1987, suffering bleeding from his throat (caused by an ulcer), and as a result was unable to work for around 3½ months. The treatment of his illness cost around Tk 30,000 as well as losses from not being able to work. He and Shireen raised the money required by selling livestock (Tk 16,900), rice and some of his wife‟s jewellery (Tk 1,600), and by taking an advance on his wages (Tk 5,500) from the printing company. This illness caused great anxiety for Rubel, and he remembers it as the most stressful period in his life to date.

In 1987, Shireen joined a BRAC cooperative (shomiti) and borrowed money which Rubel used to purchase one bigha of land. In 2000, Shireen also joined a Grameen Bank cooperative to buy further land. Since then, Shireen has taken loans nearly every year from various NGOs and they have bought land. The loans have been repaid by selling goats, shared cows and using Rubel‟s salary. The land purchased contributed to improving life at this time.

During the nationwide floods of 1988, Rubel‟s 2-year-old daughter was very ill. Medical treatment for her cost them about Tk 3000. At the end of 1988 their son was born, but in 1989 a tornado hit their village and destroyed homes and killed a number of their neighbours. They managed to escape unharmed and without much loss of property so he helped a number of his neighbours at this time. From about this time, Rubel‟s life conditions began to steadily improve with loans taken from NGOs used to purchase land. In 1997, they purchased 15 decimals of land, at a price of TK 14,000, financing the purchase by selling two shared cows, and by Shireen taking out a loan of TK 7,000 from BRAC. He brought his mother to live with him and his family, and she remained there until she died in 2006 at the age of 85, apparently from old age. Rubel also ensured that his children were being educated, both son and daughter staying in school up to year seven.


26

In 2000, Rubel purchased a one fifth share in a shallow tubewell machine at a price of TK 6,000. 2 years later he paid TK 4,000 to buy out 2 of the other partners in the machine. Later he purchased the remaining 2 shares for TK 6,000 and had full ownership of the machine. During the same period, Shireen took a loan from the Grameen Bank of TK 7,000 to purchase a sewing machine on behalf of their nephew. The nephew later repaid the cost of the sewing machine. Three years later, Rubel purchased a one thirteenth share of another shallow machine, at a price of TK 7,000. This machine is still being used. At around this time their household condition was very good, as they were growing very good rice crops (about 100 maunds from their land) and Rubel was still earning a salary in his printing job.

A marriage was arranged for Rubel‟s daughter in 2006, costing about Tk 10,000 but the agreed dowry is yet to be given as it is to cover the costs of sending the groom to work overseas. After the wedding one of the shallow machines was given to their new son-in-law.

In 2006, Rubel brought his son home from Dhaka, and enrolled him in the local school, paying TK 2,000 to purchase exercise books, clothes for school etc. However, shortly afterwards, his son went missing. Rubel estimates that his efforts to relocate and recover his son cost him around TK 5,000, and although he succeeded in bringing him safely home, the threat of the loss of his only son was enormously upsetting to him, and he was unable to work for several days after the incident. It seems that in the past some young men of Narikelbari village had disturbed Rubel‟s daughter and one of the gang wanted to marry her. When this was not possible the gang kidnapped her brother in some kind of revenge attack. Because Rubel‟s family are relative newcomers in the village they feel vulnerable to this kind of attack, especially since Rubel spends most of his time in Dhaka away from the village.

At the time of the interview, Rubel‟s only livestock holdings comprised 4 -5 hens, although he had previously owned cattle (one cow was given as part payment of the land he purchased in 1997). Rubel still works at the press in Dhaka, returning home each Friday to his family.

Rubel‟s experience illustrates how it is possible to escape childhood poverty, but that it is not

easy in the environment of rural Bangladesh. Even while opportunities exist (salaried work,

purchase of land and agricultural equipment, loans) downward pressures and shocks

(illness, river erosion, floods and violence) can quickly undermine these at any time.


27

4 Discussion

While there may be concerns about the small size and limited representativeness of the

sample of life histories we have used for this analysis, we can still be fairly confident that for

practical purposes the findings suggest a relatively high intergenerational transmission of

poverty (and wellbeing). These rates of intergenerational transmission of poverty seem to be

much higher than those detected by Pakpahan et al. (2009) in Indonesia, or those reported

by Corcoran (1995) for the United States, where for both countries fewer than half of those

poor in childhood remained poor in adulthood. This is a cause of concern for rural

Bangladesh. While poverty has been declining, for those who remain poor it seems that

poverty is still passed on from one generation with a fairly high frequency. For a more

accurate estimation of intergenerational poverty transmission, a larger panel would need to

be followed over a longer period of time. However this limited study is sufficient to raise

concerns about the need to recognise and address this kind of persistent poverty, particularly

for policy makers who aim to help provide inclusive opportunities for escape from poverty.

Hindrances to escape from intergenerational poverty seem to be due to a combination of

exposure to downward pressures, such as illness, litigation, floods, cyclones and storms, but

also due to barriers to gaining from opportunity. Key sources of opportunity seem to be found

in the form of business income, accumulation of land, livestock production, having sons or

daughters working, and income from salaried work. However, many of the chronically poor

lack the assets that enable success in business or agriculture and lack the human and social

capital which enable entry into salaried work.

Sources of opportunity which seem to have been protecting the „moving down‟ and

„chronically poor‟ groups more than the others were in the form of day labour and official

programmes such as the Vulnerable Groups Development (VGD) and Primary Education

Stipend (PES) programmes. However since the frequency of citation of these is lower among

the „move up‟ and „chronically rich‟ groups, we could conclude that these sources of

opportunity are probably more protective than promotional, and thus less effective in

permanently lifting people out of poverty. It could be argued that the amounts of money

involved in the VGD and the PES schemes are so small that for the „move up‟ and

„chronically rich‟ groups, they are dwarfed by other sources of income while for the

chronically poor, they provide the starting point for a movement out of poverty. However if

this were the case then we would expect higher frequencies of citation in the „move up‟ group

– as these are the ones who would have had that start out of poverty. The general

impression gained during the interviews was that for most VGD and PES respondents receipt

of these benefits was valued, but not sufficient to make a substantial difference in poverty

status.

In order to reduce high rate intergenerational poverty in rural Bangladesh, more effective

interventions are needed to provide the means for the poorest to seize common


28

opportunities, and also to provide better protection from the most frequent downward

pressures identified. Unless this takes place, it seems likely that many of the poorest rural

Bangladeshis will continue to pass on poverty to their children. Priority should be given to

promoting entry for the poorest into the most effective forms of opportunity which involve the

accumulation of income generating assets or employment, and at the same time provide

better protection from crises such as those associated with ill-health, dowry, death of family

members, litigation and the other common causes of crisis identified.


29

References

Ahmed, A. (2005). „Comparing Food and Cash Incentives for Schooling in Bangladesh. Study commissioned by the United Nations University‟. Washington, D.C.: International Food Policy Research Institute.

Baulch, B. (2010). „The medium-term impact of the primary education stipend in rural Bangladesh‟. Discussion Paper 976. Washington, D.C.: International Food Policy Research Institute.

Corcoran, M. (1995). „Rags to Rags: Poverty and Mobility in the United States.‟ Annual Review of Sociology, 21, 237-267.

Davis, P. (2007). „Discussions among the poor: Exploring poverty dynamics with focus groups in Bangladesh‟. Chronic Poverty Research Centre Working Paper no. 84. Manchester, UK: Chronic Poverty Research Centre.

Davis, P., and B. Baulch. (2009). „Parallel realities: Exploring poverty dynamics using mixed methods in rural Bangladesh‟. Chronic Poverty Research Centre Working Paper 142. Manchester, UK: Chronic Poverty Research Centre.

Davis, P., and B. Baulch. (2010). „Casting the net wide and deep: Lessons learned in a mixed methods study of poverty dynamics in rural Bangladesh’. Chronic Poverty Research Centre Working Paper 155. Manchester, UK: Chronic Poverty Research Centre.

Davis, P. (2011). „Vulnerability in Rural Bangladesh: Learning from life-history interviews‟. Chronic Poverty Research Centre Working Paper no. 196. Manchester, U.K.: Chronic Poverty Research Centre.

Hallman, K., Lewis, D. and Begum, S. (2007). „Assessing the impact of vegetable and fishpond technologies on poverty in rural Bangladesh‟, in: M. Adato & R. Meinzen-Dick (eds) Agricultural Research, Livelihoods, and Poverty: Studies of Economic and Social Impacts in Six Countries. Baltimore: Johns Hopkins University Press.

Kumar, N. and Quisumbing, A. (2009). „Access, Adoption, and Diffusion: Understanding the Long-term Impacts of Improved Vegetable and Fish Technologies in Bangladesh. Discussion Paper 995. Washington, D.C.: International Food Policy Research Institute.

Pakpahan, Y. M., Suryadarma, D. and Suryahadi, A. (2009). ‟Destined for destitution: intergenerational poverty persistence in Indonesia‟. CPRC Working Paper 134. Manchester, UK: Chronic Poverty Research Centre.

Sen, B. and Hulme, D. (2006). „Chronic Poverty in Bangladesh: Tales of Ascent, Descent, Marginality and Persistence‟. Dhaka and Manchester: Bangladesh Institute of Development Studies/Chronic Poverty Research Centre.

Zeller, M., Sharma, H., Ahmed, A. and Rashid, S. (2001). „Group-based Financial Institutions for the Rural Poor in Bangladesh‟. IFPRI Research Report Series, No. 120, International Food Policy Research Institute.

The Chronic Poverty Research Centre (CPRC) is an international partnership of universities, research institutes and NGOs, with the central aim of creating knowledge that contributes to both the speed and quality of poverty reduction, and a focus on assisting those who are trapped in poverty, particularly in sub-Saharan Africa and South Asia.

Partners: Bangladesh Institute of Development Studies (BIDS), Bangladesh CEDRES, University of Ouagadougou, Burkina Faso Development Initiatives, UK Development Research and Training, Uganda Economic Policy Research Center, Uganda HelpAge International, UK Indian Institute of Public Administration, India IED Afrique, Senegal Overseas Development Institute, UK Programme for Land and Agrarian Studies, South Africa Unversité Cheik Anta Diop, Senegal University Abdou Moumouni, Niger University of Legon, Ghana University of Manchester, UK University of Sussex, UK Contact: [email protected] © Chronic Poverty Research Centre 2011

www.chronicpoverty.org

passing on poverty

Documents